The SNMP exporter uses UDP for transport, so telnet to 161 isn't going to help much. I've seen very few devices implement RFC 3430.
There are usually two problems people run into. * The community string is wrong (defaults to "public") * The device has a management IP control and won't allow traffic. SNMP implementations generally don't reject incorrect authentication, rather they simply drop the packets. So it can be hard to diagnose. On Fri, Aug 21, 2020 at 2:52 PM Mario Pranjic <[email protected]> wrote: > Hi Ben, > > I ran another VM and used roles: > - https://github.com/cloudalchemy/ansible-snmp-exporter > - https://github.com/cloudalchemy/ansible-prometheus > > along with Grafana. Prometheus, node.exporter and Grafana work as intended > (as before). > > However, when I try to enable snmp_exporter: > - job_name: snmp > metrics_path: /snmp > params: > module: > - if_mib > relabel_configs: > - source_labels: > - __address__ > target_label: __param_target > - source_labels: > - __param_target > target_label: instance > - replacement: 127.0.0.1:9116 > target_label: __address__ > static_configs: > - targets: > - 192.168.10.187 > > I get error in syslog: > Aug 21 14:43:23 monnew snmp_exporter[363]: level=info > ts=2020-08-21T12:43:23.189Z caller=collector.go:224 module=if_mib > target=192.168.10.187 msg="Error scraping target" err="scrape canceled > (possible timeout) getting target 192.168.10.187" > > snmp is enabled on device on 192.168.10.187. I tried snmp version 2c and 3 > but the error message is the same. I can connect on that IP:port from > prometheus/snmp_exporter node: > root@monnew:~# telnet 192.168.10.187 161 > Trying 192.168.10.187... > Connected to 192.168.10.187. > Escape character is '^]'. > > > Port 9116 is opened on localhost on prometheus/snmp_exporter node: > root@monnew:/etc/snmp_exporter# netstat -na|grep 9116 > tcp 0 0 127.0.0.1:35274 127.0.0.1:9116 > ESTABLISHED > tcp6 0 0 :::9116 :::* LISTEN > tcp6 0 0 127.0.0.1:9116 127.0.0.1:35274 > ESTABLISHED > root@monnew:/etc/snmp_exporter# telnet 127.0.0.1 9116 > Trying 127.0.0.1... > Connected to 127.0.0.1. > Escape character is '^]'. > > What I am missing here is that generator thing. I didn't install any mibs > in this case, and, from what I see in Ansible role, no mibs or generator > has been set/run. > > Instead, snmp.yml file is being pulled from remote location: > "https://raw.githubusercontent.com/prometheus/snmp_exporter/v{{ > snmp_exporter_version }}/snmp.yml" > > Could you shed some light here, because I am obviously missing some step > along the way. > > Thanks in advance! > > Best regards, > > Mario. > > > fre. 21. aug. 2020 kl. 11:05 skrev Mario Pranjic <[email protected] > >: > >> Hm. >> >> I wrote the roles myself for Prometheus and Grafana but I used debian >> packages for Prometheus and external repository for Grafana. >> >> I am fully aware Debian often has older versions of packages (sometimes >> pretty old). To be able to use prometheus repository (as Grafana provides >> one), would make our lives a lot easier. :) >> >> I will look into these and see if things are better with newer versions. >> >> Thanks for the tip! >> >> Best regards, >> >> Mario. >> >> fre. 21. aug. 2020 kl. 10:42 skrev Ben Kochie <[email protected]>: >> >>> I highly recommend against using the Debian packages, as they are >>> extremely out of date. >>> >>> If you're using Ansible, you can use these high quality roles. >>> >>> https://github.com/cloudalchemy >>> >>> For example, https://github.com/cloudalchemy/ansible-prometheus >>> >>> We are planning an official Deb repo, but it's not ready yet. >>> >>> For the generator, it's going to be one thing you do need to build, >>> because it needs to be compiled with a distro specific net-snmp dev package. >>> >>> On Fri, Aug 21, 2020, 10:37 Mario Pranjic <[email protected]> >>> wrote: >>> >>>> Hi Ben, >>>> I use versions available in Debian repository: >>>> ii prometheus 2.7.1+ds-3+b11 >>>> amd64 Monitoring system and time series database >>>> ii prometheus-alertmanager 0.15.3+ds-3+b1 >>>> amd64 Handle and deliver alerts created by Prometheus >>>> ii prometheus-node-exporter 0.17.0+ds-3+b11 >>>> amd64 Prometheus exporter for machine metrics >>>> ii prometheus-snmp-exporter 0.14.0+ds-1+b10 >>>> amd64 Prometheus exporter for SNMP-enabled devices >>>> >>>> I try building as little as possible since all provisioning is handled >>>> by Ansible at the end. >>>> >>>> If there is prometheus external repository for Debian, that could be >>>> worth checking, but I didn't find one. >>>> >>>> >>>> fre. 21. aug. 2020 kl. 10:25 skrev Ben Kochie <[email protected]>: >>>> >>>>> What version is this, it's likely out of date. I would recommend >>>>> building from source. >>>>> >>>>> On Thu, Aug 20, 2020, 21:11 Mario Polario <[email protected]> >>>>> wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> I am trying to set-up prometheus-snmp-exporter to collect data from >>>>>> my Synology NAS. >>>>>> I downloaded mibs (Synology mibs included) to /usr/share/snmp/mibs. >>>>>> >>>>>> When running prometheus-snmp-generator I get an error: >>>>>> >>>>>> root@monitoring:/etc/prometheus# prometheus-snmp-generator generate >>>>>> >>>>>> INFO[0000] Loading MIBs from >>>>>> $HOME/.snmp/mibs:/usr/share/snmp/mibs:/usr/share/snmp/mibs/iana:/usr/share/snmp/mibs/ietf:/usr/share/mibs/site:/usr/share/snmp/mibs:/usr/share/mibs/iana:/usr/share/mibs/ietf:/usr/share/mibs/netsnmp >>>>>> source="net_snmp.go:136" >>>>>> WARN[0000] NetSNMP reported 2 parse errors >>>>>> source="main.go:103" >>>>>> FATA[0000] Error parsing yml config: Invalid metric type override >>>>>> 'EnumAsInfo' source="main.go:44" >>>>>> >>>>>> >>>>>> I tried removing all from generator.yml except Synology module, but >>>>>> no change. >>>>>> >>>>>> I was following instructions noted >>>>>> in /usr/share/doc/prometheus-snmp-exporter/README.Debian but I can't seem >>>>>> to generate snmp.yml file. >>>>>> >>>>>> Any help in understanding what I did wrong is more than welcomed. >>>>>> >>>>>> Thanks in advance! >>>>>> >>>>>> Best regards, >>>>>> >>>>>> Mario. >>>>>> >>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "Prometheus Users" group. >>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>> send an email to [email protected]. >>>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/prometheus-users/b8f2f195-efc1-4c48-b85b-88a8687964bfn%40googlegroups.com >>>>>> <https://groups.google.com/d/msgid/prometheus-users/b8f2f195-efc1-4c48-b85b-88a8687964bfn%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>> . >>>>>> >>>>> -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/CABbyFmpjF8JnvTpffXDbX0nTY9SRnBoWtB96dv57QB%3DhERUmhA%40mail.gmail.com.
